[M-IS06] Dust

[MIS06-04] What are the main land-surface processes affecting saltation activities in a dust hotspot of the Mongolian desert steppe?

*Kaman Kong¹, Banzragch Nandintsetseg^1,2, Masato Shinoda¹, Masahide Ishizuka³, Yasunori Kurosaki⁴ (1.Graduate School of Environmental Studies, Nagoya University, 2.School of Arts and Sciences, National University of Mongolia, 3.Faculty of Engineering and Design, Kagawa University, 4.Arid Land Research Center, Tottori University)

Keywords:Asian dust, Dead-leaf hypothesis, Ecosystem model, Freeze-thaw process, Saltation, Threshold wind speed

Wind erosion is a global phenomenon that occurs in arid and semi-arid regions worldwide, including the Mongolian desert steppe. In recent decades, dust outbreaks increased in the Mongolian desert steppe posing a serious threat to local ecosystem and animal/human health in both dust source and downwind regions. Dust outbreaks are impacted by wind speed (erosivity factor) as well as by land surface conditions (erodibility factor), such as vegetation cover, freeze-thaw process, soil moisture, soil crust, and snow cover, because they affect threshold wind speed for wind emission. However, in the Mongolian desert steppe, these physical relationships between surface conditions and dust outbreaks have not fully been investigated due to constraint of field observation. In this study, we integrated field observations and process-based ecosystem model DAYCENT to explore the effects of land-surface processes on dust outbreaks and to understand the relative contributions of erosivity and erodibility for wind erosion at Tsogt-Ovoo on the Mongolian desert steppe during 2012–2017. The ecosystem model gave reasonably good simulations of the vegetation, and soil moisture and temperature dynamics. Observational results show that dust outbreaks were mainly caused by westerly and northerly strong winds (>10 m/s) for late February to August. That period can be divided into primary (late February – early May) and secondary sub-periods (late May – August). Based on multiple regression analysis, we found that the threshold wind speed did not correlate with soil moisture, but decreased with increasing soil temperature in the primary sub-period (r=-0.65, p<0.01) and increased with increasing vegetation components of standing dead (dead-leaf hypothesis) and live grasses in the secondary sub-period (r=0.87, p<0.05). These variations of threshold wind speed possibly related to the snow cover and freeze-thaw processes in the primary sub-period, and dead and live grasses in the secondary sub-period, respectively. Our novel model of threshold wind speed explained that more frequent and higher saltation flux in 2012 and 2015 resulted from a combination of frequent strong winds and a lower threshold wind speed for dust emission, which was related to higher soil temperature and smaller vegetation components. In addition, other conditions (e.g. soil physical crust formation and soil particle size distribution) likely affected dust occurrences, particular during transitional period in May.